Analog-to-digital converters (A/D converters or ADCs) find common use in a wide variety of products that operate in both the analog and digital domains. To illustrate, wireless devices, such as cellular phones, laptop computers and personal digital assistants, typically use A/D converters to convert wireless analog signals that carry information into digital information suitable for processing and manipulation by microprocessors and other digital components.
While often critical to proper operation, A/D converters typically consume considerable power, a situation that is particularly relevant to devices that have a limited supply of power, such as battery-powered devices or portable devices. The power consumption of an A/D converter typically is related to the dynamic range of the A/D converter (i.e., the number of active output bits of the A/D converter). Accordingly, one common technique to reduce power consumption is to use an A/D converter with a limited dynamic range. However, in many applications, the implementation of an A/D converter having a limited dynamic range often requires additional front-end signal processing to condition the input signal for conversion by the limited dynamic-range A/D converter. For example, in radio frequency (RF) applications, the use of a limited dynamic-range A/D converter typically requires that increased selectivity be provided by the RF front end, thereby increasing the cost and complexity of the RF front end. Moreover, the power consumption of the additional circuitry required to implement the more complex RF front end typically offsets or even exceeds the power savings achieved by using a limited dynamic-range A/D converter.
Due to the additional cost and complexity of implementing limited dynamic-range A/D converters, another conventional technique has been developed whereby the dynamic range of an A/D converter is adjusted in response to a long-term averaging of the digital signal output by the A/D converter. This long-term averaging technique implements circuits to square, sum and filter the on-channel signal content at the back end of a RF transceiver, where the decision on whether to adjust the dynamic range of the A/D converter is made after considerable processing and delay. This technique requires considerable circuitry to implement the squaring, summing and filtering operations, which significantly increases the complexity of the RF application as well as its power consumption. Moreover, this technique is reliant on a significant amount of prior signal history due to the squaring, summing and filtering operations to average the signal at the back end and therefore introduces a significant delay between when a condition occurs and when this condition is identified and appropriate adjustment of the A/D converter is made, which may result in clipping and therefore requiring a change in the dynamic range of the A/D converter. Accordingly, an improved technique for adjusting the dynamic range of an A/D converter would be advantageous.
The use of the same reference symbols in different drawings indicates similar or identical items.